The overall goal of this project is to define the cerebral response to insulin-induced hypoglycemia. In diabetes, hypoglycemia limits our ability to achieve normoglycemia and thereby reduce the long-term complications of the disease. Recurrent episodes of hypoglycemia also lead to the clinical syndrome of hypoglycemia unawareness in which patients lose their ability to detect hypoglycemia until neuroglycopenia occurs. How the brain adapts to recurrent hypoglycemia is uncertain, but greater understanding of this process will provide insights into how we can better treat patients with diabetes. In this investigation, we will directly examine how the human brain responds to hypoglycemia in vivo using high field magnetic resonance imaging and 13C magnet resonance spectroscopy. We will take advantage of the relationship between cerebral blood flow and neuronal activation to determine if glucose sensing regions can be identified in humans and to determine if the response of the brain to hypoglycemia is altered by antecedent glycemia in diabetes. We will also determine if insulin has effects on the brain that are independent of glycemia. The hypotheses we intend to examine and the aims of the studies planned to address each hypothesis are below. Hypothesis #1: During hypoglycemia, regions of the brain with an abundance of glucose sensing neurons such as the hypothalamus, amygdala, and brainstem will increase blood flow at a higher glucose concentration than will other brain regions. Aim #1: To determine blood and brain glucose concentrations at which regional cerebral blood flow is increased during hypoglycemia. Hypothesis #2: Cerebral blood flow will increase during hypoglycemia at the same brain glucose but at different blood glucose concentration in patients with poorly controlled diabetes, patients with type 1 diabetes and hypoglycemia and in healthy volunteers.Aim #2A: To determine the blood and brain glucose concentrations at which regional cerebral blood flow increases during hypoglycemia in patients with poorly controlled diabetes and compare these values to those measured in healthy volunteers.Aim #2B: To determine the blood and brain glucose concentrations at which regional cerebral blood flow increases during hypoglycemia in patients with type 1 diabetes and hypoglycemia unawareness and compare these values to those measured in healthy volunteers. Hypothesis #3: Insulin has effects independent of glucose on cerebral blood flow.Aim #3: To determine if cerebral blood flow is altered by changes in serum insulin concentrations. Aim #4: To determine if insulin alters neuronal activation as measured event-related potentials.